popov/cpuminer-opt-gpu
1
0
Fork 0
mirror of https://github.com/JayDDee/cpuminer-opt.git synced 2025-09-17 23:44:27 +00:00
Code Issues Packages Projects Releases Wiki Activity

v3.11.0

Browse Source
This commit is contained in:
Jay D Dee
2020-01-02 23:54:08 -05:00
parent 241bc26767
commit 3572cb53c4
118 changed files with 7030 additions and 1575 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
algo/lyra2/allium-4way.c
View File

@@ -55,7 +55,6 @@ void allium_8way_hash( void *state, const void *input )
dintrlv_8x64( hash0, hash1, hash2, hash3, hash4, hash5, hash6, hash7,
vhash, 256 );
intrlv_2x256( vhash, hash0, hash1, 256 );
LYRA2RE_2WAY( vhash, 32, vhash, 32, 1, 8, 8 );
dintrlv_2x256( hash0, hash1, vhash, 256 );
@@ -69,19 +68,6 @@ void allium_8way_hash( void *state, const void *input )
LYRA2RE_2WAY( vhash, 32, vhash, 32, 1, 8, 8 );
dintrlv_2x256( hash6, hash7, vhash, 256 );
/*
LYRA2RE( hash0, 32, hash0, 32, hash0, 32, 1, 8, 8 );
LYRA2RE( hash1, 32, hash1, 32, hash1, 32, 1, 8, 8 );
LYRA2RE( hash2, 32, hash2, 32, hash2, 32, 1, 8, 8 );
LYRA2RE( hash3, 32, hash3, 32, hash3, 32, 1, 8, 8 );
LYRA2RE( hash4, 32, hash4, 32, hash4, 32, 1, 8, 8 );
LYRA2RE( hash5, 32, hash5, 32, hash5, 32, 1, 8, 8 );
LYRA2RE( hash6, 32, hash6, 32, hash6, 32, 1, 8, 8 );
LYRA2RE( hash7, 32, hash7, 32, hash7, 32, 1, 8, 8 );
*/
intrlv_4x128( vhashA, hash0, hash1, hash2, hash3, 256 );
intrlv_4x128( vhashB, hash4, hash5, hash6, hash7, 256 );
@@ -105,20 +91,6 @@ void allium_8way_hash( void *state, const void *input )
LYRA2RE_2WAY( vhash, 32, vhash, 32, 1, 8, 8 );
dintrlv_2x256( hash6, hash7, vhash, 256 );
/*
LYRA2RE( hash0, 32, hash0, 32, hash0, 32, 1, 8, 8 );
LYRA2RE( hash1, 32, hash1, 32, hash1, 32, 1, 8, 8 );
LYRA2RE( hash2, 32, hash2, 32, hash2, 32, 1, 8, 8 );
LYRA2RE( hash3, 32, hash3, 32, hash3, 32, 1, 8, 8 );
LYRA2RE( hash4, 32, hash4, 32, hash4, 32, 1, 8, 8 );
LYRA2RE( hash5, 32, hash5, 32, hash5, 32, 1, 8, 8 );
LYRA2RE( hash6, 32, hash6, 32, hash6, 32, 1, 8, 8 );
LYRA2RE( hash7, 32, hash7, 32, hash7, 32, 1, 8, 8 );
*/
intrlv_8x64( vhash, hash0, hash1, hash2, hash3, hash4, hash5, hash6,
hash7, 256 );
@@ -232,11 +204,11 @@ void allium_4way_hash( void *state, const void *input )
allium_4way_ctx_holder ctx __attribute__ ((aligned (64)));
memcpy( &ctx, &allium_4way_ctx, sizeof(allium_4way_ctx) );
blake256_4way( &ctx.blake, input + (64<<2), 16 );
blake256_4way_update( &ctx.blake, input + (64<<2), 16 );
blake256_4way_close( &ctx.blake, vhash32 );
rintrlv_4x32_4x64( vhash64, vhash32, 256 );
keccak256_4way( &ctx.keccak, vhash64, 32 );
keccak256_4way_update( &ctx.keccak, vhash64, 32 );
keccak256_4way_close( &ctx.keccak, vhash64 );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash64, 256 );
@@ -261,7 +233,7 @@ void allium_4way_hash( void *state, const void *input )
intrlv_4x64( vhash64, hash0, hash1, hash2, hash3, 256 );
skein256_4way( &ctx.skein, vhash64, 32 );
skein256_4way_update( &ctx.skein, vhash64, 32 );
skein256_4way_close( &ctx.skein, vhash64 );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash64, 256 );

2
algo/lyra2/lyra2-gate.c
View File

@@ -229,7 +229,7 @@ void phi2_build_extraheader( struct work* g_work, struct stratum_ctx* sctx )
bool register_phi2_algo( algo_gate_t* gate )
{
// init_phi2_ctx();
gate->optimizations = SSE2_OPT | AES_OPT | SSE42_OPT | AVX2_OPT;
gate->optimizations = SSE2_OPT | AES_OPT | SSE42_OPT | AVX2_OPT | AVX512_OPT;
gate->get_work_data_size = (void*)&phi2_get_work_data_size;
gate->decode_extra_data = (void*)&phi2_decode_extra_data;
gate->build_extraheader = (void*)&phi2_build_extraheader;

2
algo/lyra2/lyra2h-4way.c
View File

@@ -20,7 +20,7 @@ static __thread blake256_4way_context l2h_4way_blake_mid;
void lyra2h_4way_midstate( const void* input )
{
blake256_4way_init( &l2h_4way_blake_mid );
blake256_4way( &l2h_4way_blake_mid, input, 64 );
blake256_4way_update( &l2h_4way_blake_mid, input, 64 );
}
void lyra2h_4way_hash( void *state, const void *input )

12
algo/lyra2/lyra2rev2-4way.c
View File

@@ -44,7 +44,7 @@ void lyra2rev2_8way_hash( void *state, const void *input )
lyra2v2_8way_ctx_holder ctx __attribute__ ((aligned (64)));
memcpy( &ctx, &l2v2_8way_ctx, sizeof(l2v2_8way_ctx) );
blake256_8way( &ctx.blake, input + (64<<3), 16 );
blake256_8way_update( &ctx.blake, input + (64<<3), 16 );
blake256_8way_close( &ctx.blake, vhash );
rintrlv_8x32_8x64( vhashA, vhash, 256 );
@@ -176,12 +176,12 @@ void lyra2rev2_4way_hash( void *state, const void *input )
lyra2v2_4way_ctx_holder ctx __attribute__ ((aligned (64)));
memcpy( &ctx, &l2v2_4way_ctx, sizeof(l2v2_4way_ctx) );
blake256_4way( &ctx.blake, input + (64<<2), 16 );
blake256_4way_update( &ctx.blake, input + (64<<2), 16 );
blake256_4way_close( &ctx.blake, vhash );
rintrlv_4x32_4x64( vhash64, vhash, 256 );
keccak256_4way( &ctx.keccak, vhash64, 32 );
keccak256_4way_update( &ctx.keccak, vhash64, 32 );
keccak256_4way_close( &ctx.keccak, vhash64 );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash64, 256 );
@@ -201,7 +201,7 @@ void lyra2rev2_4way_hash( void *state, const void *input )
intrlv_4x64( vhash64, hash0, hash1, hash2, hash3, 256 );
skein256_4way( &ctx.skein, vhash64, 32 );
skein256_4way_update( &ctx.skein, vhash64, 32 );
skein256_4way_close( &ctx.skein, vhash64 );
dintrlv_4x64( hash0, hash1, hash2, hash3, vhash64, 256 );
@@ -217,7 +217,7 @@ void lyra2rev2_4way_hash( void *state, const void *input )
intrlv_4x32( vhash, hash0, hash1, hash2, hash3, 256 );
bmw256_4way( &ctx.bmw, vhash, 32 );
bmw256_4way_update( &ctx.bmw, vhash, 32 );
bmw256_4way_close( &ctx.bmw, state );
}
@@ -242,7 +242,7 @@ int scanhash_lyra2rev2_4way( struct work *work, uint32_t max_nonce,
mm128_bswap32_intrlv80_4x32( vdata, pdata );
blake256_4way_init( &l2v2_4way_ctx.blake );
blake256_4way( &l2v2_4way_ctx.blake, vdata, 64 );
blake256_4way_update( &l2v2_4way_ctx.blake, vdata, 64 );
do
{

13
algo/lyra2/lyra2rev3-4way.c
View File

@@ -209,7 +209,7 @@ void lyra2rev3_8way_hash( void *state, const void *input )
lyra2v3_8way_ctx_holder ctx __attribute__ ((aligned (64)));
memcpy( &ctx, &l2v3_8way_ctx, sizeof(l2v3_8way_ctx) );
blake256_8way( &ctx.blake, input + (64*8), 16 );
blake256_8way_update( &ctx.blake, input + (64*8), 16 );
blake256_8way_close( &ctx.blake, vhash );
dintrlv_8x32( hash0, hash1, hash2, hash3,
@@ -252,7 +252,7 @@ void lyra2rev3_8way_hash( void *state, const void *input )
intrlv_8x32( vhash, hash0, hash1, hash2, hash3,
hash4, hash5, hash6, hash7, 256 );
bmw256_8way( &ctx.bmw, vhash, 32 );
bmw256_8way_update( &ctx.bmw, vhash, 32 );
bmw256_8way_close( &ctx.bmw, state );
}
@@ -277,7 +277,7 @@ int scanhash_lyra2rev3_8way( struct work *work, const uint32_t max_nonce,
mm256_bswap32_intrlv80_8x32( vdata, pdata );
blake256_8way_init( &l2v3_8way_ctx.blake );
blake256_8way( &l2v3_8way_ctx.blake, vdata, 64 );
blake256_8way_update( &l2v3_8way_ctx.blake, vdata, 64 );
do
{
@@ -334,8 +334,7 @@ void lyra2rev3_4way_hash( void *state, const void *input )
lyra2v3_4way_ctx_holder ctx __attribute__ ((aligned (64)));
memcpy( &ctx, &l2v3_4way_ctx, sizeof(l2v3_4way_ctx) );
// blake256_4way( &ctx.blake, input, 80 );
blake256_4way( &ctx.blake, input + (64*4), 16 );
blake256_4way_update( &ctx.blake, input + (64*4), 16 );
blake256_4way_close( &ctx.blake, vhash );
dintrlv_4x32( hash0, hash1, hash2, hash3, vhash, 256 );
@@ -358,7 +357,7 @@ void lyra2rev3_4way_hash( void *state, const void *input )
LYRA2REV3( l2v3_wholeMatrix, hash3, 32, hash3, 32, hash3, 32, 1, 4, 4 );
intrlv_4x32( vhash, hash0, hash1, hash2, hash3, 256 );
bmw256_4way( &ctx.bmw, vhash, 32 );
bmw256_4way_update( &ctx.bmw, vhash, 32 );
bmw256_4way_close( &ctx.bmw, state );
}
@@ -383,7 +382,7 @@ int scanhash_lyra2rev3_4way( struct work *work, const uint32_t max_nonce,
mm128_bswap32_intrlv80_4x32( vdata, pdata );
blake256_4way_init( &l2v3_4way_ctx.blake );
blake256_4way( &l2v3_4way_ctx.blake, vdata, 64 );
blake256_4way_update( &l2v3_4way_ctx.blake, vdata, 64 );
do
{

8
algo/lyra2/lyra2z-4way.c
View File

@@ -149,7 +149,7 @@ static __thread blake256_8way_context l2z_8way_blake_mid;
void lyra2z_8way_midstate( const void* input )
{
blake256_8way_init( &l2z_8way_blake_mid );
blake256_8way( &l2z_8way_blake_mid, input, 64 );
blake256_8way_update( &l2z_8way_blake_mid, input, 64 );
}
void lyra2z_8way_hash( void *state, const void *input )
@@ -166,7 +166,7 @@ void lyra2z_8way_hash( void *state, const void *input )
blake256_8way_context ctx_blake __attribute__ ((aligned (64)));
memcpy( &ctx_blake, &l2z_8way_blake_mid, sizeof l2z_8way_blake_mid );
blake256_8way( &ctx_blake, input + (64*8), 16 );
blake256_8way_update( &ctx_blake, input + (64*8), 16 );
blake256_8way_close( &ctx_blake, vhash );
dintrlv_8x32( hash0, hash1, hash2, hash3,
@@ -247,7 +247,7 @@ static __thread blake256_4way_context l2z_4way_blake_mid;
void lyra2z_4way_midstate( const void* input )
{
blake256_4way_init( &l2z_4way_blake_mid );
blake256_4way( &l2z_4way_blake_mid, input, 64 );
blake256_4way_update( &l2z_4way_blake_mid, input, 64 );
}
void lyra2z_4way_hash( void *state, const void *input )
@@ -260,7 +260,7 @@ void lyra2z_4way_hash( void *state, const void *input )
blake256_4way_context ctx_blake __attribute__ ((aligned (64)));
memcpy( &ctx_blake, &l2z_4way_blake_mid, sizeof l2z_4way_blake_mid );
blake256_4way( &ctx_blake, input + (64*4), 16 );
blake256_4way_update( &ctx_blake, input + (64*4), 16 );
blake256_4way_close( &ctx_blake, vhash );
dintrlv_4x32( hash0, hash1, hash2, hash3, vhash, 256 );